For proper licensing of open source software, it is necessary to know whether the files form a combined work and, if so, whether at least one file of such a combined work may only be copied and distributed under the terms of a copyleft license. If all files are binary files and contain the so-called ELF data structure, this structure can be used to obtain information about which symbols are undefined and require an external file to be resolved and imported from it. The command line tool callgraph was developed for this purpose; it reads all files in specified directories, examines their ELF headers, and draws a recursive dependency graph of the files found. Originally written by Armijn Hemel, the callgraph tool has been extended by OSADL and made available to the public in an OSADL Git repository. A more detailed description and some sample graphs can be found here.

When a new version of a software package for which a final license clearing has already been performed is released and the associated product needs to be updated, the license clearing must be performed again. However, often only a small number of files are updated. Therefore, it would be helpful if there were a software that examines all the old and new files, check which ones have been changed or added, and copy these disjunct files to a new directory. This would make it possible to limit the work to the changed or added files in the new directory and reuse the licensing data of the remaining files that have not been changed. One such program is the Python script Disjunctify.py which is publicly available in the OSADL Git repository and described in more detail here.

Most Open Source licenses require that the license text and copyright notices be delivered to the recipient along with the software. This undoubtedly also applies in the event that Open Source software is copied to a user's computer when a web page is loaded and executed locally, as is regularly the case with software written in JavaScript, for example. Since it is not evident how this is best achieved, this application note describes a possible implementation and contains a practical realization.